1 Demographic Transition, Demographic Dividend, and Lewis Turning Point in China Cai Fang Institute of Population and Labor Economics, CASS Abstract: The disagreements on changed stages of demographic transition and the role of a demographic dividend in a dual economy development process often lead to wide debates among scholars about China’ s development stages. This paper tries to reveal the nexus between demographic transition and dual economy development: the common starting point, close related processes, and identical characteristics of stages. Based on the empirical evidence of population dynamics, the paper supports the judgment of diminishing demographic dividends and an imminent Lewis turning point in China. It also argues that further economic growth and thus faster entry into a high-income economy is the key and only way to close the “aging before affluence” gap. Accordingly, the paper concludes by proposing measures to exploit the potential of the first demographic dividend, creating conditions for a second demographic dividend, and tapping new sources of economic growth. Key words: Demographic transition; Demographic Dividend; Lewis turning point; Aging before affluence JEL Classification: J11, O53 1. Introduction One of the hot topics that inspire debates among scholars, policy researchers, and even policy-makers is whether Chinese economic growth is losing its source and momentum from demographic dividends. Another related topic is whether China reaches its Lewis turning
22
Embed
Demographic Transition, Demographic Dividend, and Lewis ...
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
1
Demographic Transition, Demographic Dividend, and
Lewis Turning Point in China
Cai Fang
Institute of Population and Labor Economics, CASS
Abstract: The disagreements on changed stages of demographic transition and the role of a
demographic dividend in a dual economy development process often lead to wide debates
among scholars about China’s development stages. This paper tries to reveal the nexus
between demographic transition and dual economy development: the common starting point,
close related processes, and identical characteristics of stages. Based on the empirical
evidence of population dynamics, the paper supports the judgment of diminishing
demographic dividends and an imminent Lewis turning point in China. It also argues that
further economic growth and thus faster entry into a high-income economy is the key and
only way to close the “aging before affluence” gap. Accordingly, the paper concludes by
proposing measures to exploit the potential of the first demographic dividend, creating
conditions for a second demographic dividend, and tapping new sources of economic growth.
Key words: Demographic transition; Demographic Dividend; Lewis turning point; Aging
before affluence
JEL Classification: J11, O53
1. Introduction
One of the hot topics that inspire debates among scholars, policy researchers, and even
policy-makers is whether Chinese economic growth is losing its source and momentum from
demographic dividends. Another related topic is whether China reaches its Lewis turning
2
point – namely, labor supply is no longer unlimited. In a research paper, Cai and Wang (2005)
estimates that the decline in the population dependence ratio, as a proxy for demographic
dividend, contributed 26.8 percent to per capita GDP growth during 1982 to 2000, and warns
that such a demographic dividend will disappear after 2013 when dependence ratio stops
decreasing and begins increasing. By examining changes in population age structure, labor
demand trends, widespread labor shortages, wage rises of ordinary workers, Cai (2008a,
2008b) asserts that the Lewis turning point has indeed come, and points out its policy
implications in terms of growth pattern transformation, income distribution trends, labor
market institutional construction and human capital accumulation.
While some researchers and even some policy documents support and cite the conclusion
about the arrival of the Lewis turning point, others strongly disagree. In previous replies, Cai
(2008a) tried to provide a wide range of evidence to defend his arguments. It turns out that
people come out with conflicting opinions about economic reality, and explain the same
phenomenon in different ways. Many still firmly hold to the conventional wisdom that there
is a massive and increasing working age population and thus endless surplus labor force in
rural areas, and that this is an unchangeable characteristic of China. Propositions which assert
the possibility of labor shortage or disappearance of the surplus labor force in agriculture –
namely that the Lewis turning point is arriving - are not widely agreed1. Specifically, all
skeptical and critical comments on the judgment about an ongoing Lewis turning point, which
results from a static understanding of population and labor force in China, are generally
puzzled by the Chinese statistics. In what follows, I unveil some aspects of such statistical
puzzles.
First, given that the official survey on utilization of agricultural workforce is unable to
reflect the fast changing reality of agricultural production, some scholars are unaware of the
changed situation, while others who have tried to understand the statistics are actually trapped
1 This existing paper does not intend to discuss the divarication that is caused by different definitions
of Lewis turning point. According to Lewis (1972) and Ranis and Fei (1961), Lewis turning point can
be referred to as the period of time at which expansion of labor demand exceeds that of labor supply
and, as a result, wage rate of ordinary workers starts to rise, while wage of agricultural sector is not yet
determined by its marginal productivity of labor and the difference of marginal productivity of labor
between agricultural and non-agricultural sectors remains. And the time when the wage rates in
agricultural and non-agricultural sectors are both determined by their marginal productivity of labor
and the gap in productivities disappears can be called commercial point. Only at this time, dual
economy ends.
3
in “the tyranny of numbers” as was Young (1994) when he tried to challenge the “East Asian
miracle”. Either case makes any endeavor of econometric analysis hostage to the data. The
point is that economic reform in China has been too fast for the statistical system to catch up
(Ravallion and Chen, 1999). One of the many examples that cause confusion concerns the
accurate numbers of agricultural workforce actually used. In 2008, the reported total labor
force engaged in agriculture was 307 million, accounting for 39.6 percent of the country’s
total employment, and the figure provided by 2008 Agricultural Census was even higher.
However, the actual input of labor in agricultural production, calculated based on agricultural
costs survey data, turns out to be much less than any published aggregated figures (Cai and
Wang, 2008). Taking into consideration the changing trend of working age population in rural
area, the updated situation of labor migration from rural to urban sectors, and the extent to
which agriculture is mechanised, one must conclude that the actually used workforce in
agricultural production is much less than what official statistical publications declare.
Therefore, the declaration that there is large amount of surplus labor to be shifted from
agriculture (e.g. Lau, 2010a) and the econometric estimation of marginal productivity of labor
in agriculture (e.g. Minami and Ma, 2009), which are both based on the aggregated dataset,
tend to overestimate the degree of labor surplus in agriculture and conclude that the Lewis
turning point has not come to China.
Second, scholars have difficulties in interpreting statistics on labor market and rural and
urban employment, and thus they often elicit conclusions that deviate from reality. As the
result of sectoral changes and increasing diversification of ownership, especially after the
labor market shock in the late 1990s, multifaceted sectors have appeared to absorb labor into
urban areas, contrary to the pre-reform period when state and collective sectors dominated
employment absorption. Among those sectors of employment, large scale informal
employment, as the byproduct of reemployment of the laid-off and of diversity of
employment, is new to China. Meanwhile, massive numbers of rural laborers have
transformed their jobs from agricultural to non-agricultural sectors, amounting to 240 million,
of which 145 million migrated into cities. In routine statistics, neither informal employment
of urban residents nor employment of migrant workers in urban sectors has been
authoritatively reported, except for estimated figures of migrant workers based on sampling
4
surveys and aggregated estimate of informally employed urban residents under certain
assumptions (Cai, 2004). We can view the difference between the number of total
employment based on the unit reporting system and the number of employment based on the
household survey as a proxy for urban informal employment, which amounts to 95.1 million
and accounts for 31.5 percent of total urban employment in 2008. It is however helpless if one
wants to do any statistical analysis on structural characteristics of the total employment,
because of lack of disaggregated data on it. Moreover, the statistical authority has so far not
promulgated an alternative surveyed unemployment rate data series to the discredited
registered unemployment rate, and that leads scholars to do various guesstimates on the
unemployment rate. Based on incomplete employment data and unfounded guesstimates,
Chinese and international scholars often educed conclusions such as zero growth of
employment and a high and increasing unemployment rate (Ru et al., 2008, p. 22; Rawski,
2001; Solinger, 2001) and doubt the authenticity of the widespread labor shortage.
Third, there is no officially published systematic data and up to date information on the
status of demographic change and population dynamics. While various rounds of national
population censuses provide information about population changes, due to lack of consensus
on some important parameters of China’s demographics such as the actual total fertility rate
(TFR) 1
, no authoritative projections of population change, including predictions of
magnitude and the age structure of the population, have been periodically publicized. The
public and academia therefore do not have updated information about t population
developments trends and many conceive that the peak of population growth will be reached in
or after 2040 and then the total population in China will as many as 1.6 billion (e.g. Lau,
2010b). More specifically, most scholars ignore the fact that the growth of China’s working
age population has been slowing and thus the demographic foundation of unlimited labor
supply has been shrinking, and therefore they are unwilling to accept the assertion of an
ongoing Lewis turning point associated with a diminishing demographic dividend.
1 The 5
th National Population Census conducted in 2000 shows that China’s TFR was 1.32, which is
even lower than policy allowable level of 1.51. Many doubt such a result (e.g. Yu, 2002). Since then the
debates on what is the actual TFR of China have existed among scholars and policy researchers.
Generally speaking, the government departments responsible for implementing the population control
policy tend to believe a higher TFR, whereas scholars believe a lower TFR. In spite of the
disagreement, the estimates mostly fall in the range of 1.6 to 1.8, which are all significantly lower than
the replacement level of 2.1.
5
It is obvious that an undistorted understanding of status and trends of demographic
transition will help scholars and policy researchers better understand the state of labor market
and will serve as a foundation for policy decisions on how China can sustain its economic
growth. The following sections of the paper argue that demographic transition and dual
economy development have a common starting point, related and similar characteristics of
development stages, as well as overlapping processes to a large extent, so that the
demographic window of opportunity in which a demographic dividend is obtained is one of
the stages of dual economy development. Accordingly, the theoretical and empirical work and
reasoning about a diminishing demographic dividend and incoming Lewis turning point kill
two birds with one stone. The rest of the paper is organized as follows. Section 2 reveals a
stylized fact about relationship between demographic transition and dual economy
development based on international experiences. Section 3 depicts the China’s process of
demographic transition and its impact on economic growth. Section 4 tries to answer the
question of how the “aging before affluence” gap can be narrowed. Section 5 concludes by
drawing policy implications of the issues discussed in the paper.
2. Stages of Demographic Transition and Development of Dual
Economy
The theory of dual economy coined by Lewis (1954) divides a typical developing
economy into two sectors: the agricultural and modern sectors. Because labor force is
superfluous relative to capital and land in agriculture, its marginal productivity in the sector is
very low, even as low as zero or below. As the modern sector expands surplus labor in
agriculture is transferred to modern sector without substantial rise of wages, and the whole
process is typically called the development of dual economy. Such a process continues until it
reaches a point at which the growth of labor demand succeeds growth of labor supply and
further labor transfer requires increase in unskilled workers’ wage rate. That point is generally
called Lewis turning point. In spite of its ups and downs in economics history (Ranis, 2004),
Lewisian theory of dual economy has always remained as an important theoretical model of
6
development economics.
Even before Lewis’s prominent paper first appeared, the mature form of demographic
transition theory had been published already1
. Corresponding to pre- and
post-industrialization periods, demographic transition is categorized into three stages, which
are respectively characterized by (1) high birth rate, high death rate and low natural growth
rate of population, (2) high birth rate, low death rate and high natural growth rate of
population, and (3) low birth rate, low death rate and low natural growth rate of population.
Although we cannot judge whether Lewis noticed those literatures in demography, there is no
lack of demographic assumptions related to the theory of demographic transition in his
description of development of dual economy. While defining unlimited supply of labor, the
key concept of the theory of dual economy, he explains: “unlimited supply of labor may be
said to exist in those countries where population is so large relatively to capital and natural
resources, that there are large sectors of the economy where the marginal productivity of labor
is negligible, zero, or even negative.” The connotative assumption of this statement is that a
typical dual economy characterized by unlimited supply of labor is at the second stage of
demographic transition – that is, natural growth rate of population is high as the result of
declined mortality and inertial high birth rate. Since agriculture is primary sector in the
sectoral chain, it is the first place where the abundant population and surplus labor force roost.
The key of comprehending the logical and empirical relationship between demographic
transition and development of dual economy is to explore how demographic dividend is
engendered and obtained. In early literature of demography and economics, the
population-development nexus was discussed by focusing on the relationship between
economic growth rate and population growth rate or population quantum, while the discussion
on demographic transition was mainly about demographic contents such as population
quantity, birth rate, and death rate, but not closely related to economic growth. Besides,
mainstream of growth theory, while incorporating population into endogenous growth,
usually neglects characteristics of demographic transition of dual economy. After long neglect
1 Whereas Thompson (1929) first identified the three stages of demographic transition and another
scholar added two more later stages, they were both not considered as the farther of theory of
demographic transition, because they did not provide standard theoretical explanation on decline of
fertility. The honor was later awarded to Notestein (1945). Please see Caldwell (1976) for a brief
history of this field.
7
of economic development and structural characteristics of population, particularly the relation
between population age structure and labor supply, as all developed countries and many
newly industrialized economies successively completed their demographic transition process,
demographers became conscious of population aging and its consequences. Economists
farther unveil the change in working age population going with fertility decline and its effect
on sources of economic growth (Williamson, 1997). That is, in the interval between a sooner
decline of death rate and later decline of birth rate, natural growth rate of population is usually
at its fast rise, youth dependence ratio is also increasing. After a certain period of time, as
fertility decreases and the baby boomers grow up, the proportion of working age population
enhances accordingly. The further decline in fertility as a result of economic and social
developments causes slowdown of natural growth rate of population, and the structural
consequence of such a dynamics is population aging. In short, following a reversed U shape
pattern – namely, natural growth rate of population first increases and then declines after a
turning point, with an interval of about one generation, growth rate of working age population
presents a similar pattern of changes.
During the period in which population age structure is most productive, adequate supply
of labor and high savings rate afford an extra source of economic growth and thus form
demographic dividend. Consequently, once demographic transition exceeds this stage –
namely, population age structure becomes less and less productive, because of the rapid aging,
such conventionally defined demographic dividend gradually disappears. Since the stages in
demographic transition can be sufficiently characterized by changes in total fertility rate
(TFR), one can theoretically expect the following relation between demographic transition
and economic growth (Figure 1): the stage of high TFR coincides with steady state of low
growth rate; as TFR falls and as a result, a more productive population age structure comes
into being, demographic dividend promotes economic growth to a higher rate; when TFR
further drops to low level and population ages, economic growth rate again shrinks to low
steady state. Correspondingly, at a certain stage of demographic transition when TFR declines
rapidly and population age structure becomes more and more productive, there forms a
demographic window of opportunity.
8
Figure 1 Relationship between Fertility and Economic Growth
It is worth noting that factors impacting the performance of economic growth are
multifold, not just of population. This is also true in explaining both the steady state of growth
rate of low income economies, known for poverty trap, and the steady state of growth rate of
high income economies struggling in technological innovation frontier. For example, in the
empirical works of defending neoclassical growth theory, economists have found more than
one hundred explanatory variables, which are statistically significant in unveiling
determinants of growth performances, but none is sufficient and exclusive (Sala-i-Martin,
1997). For simplicity, we also put aside the retroaction effect of economic growth on
demographic transition1 and focus on the straightforward relationship between fertility and
economic growth. Under the assumption made above, this fertility-growth nexus can be
deduced from the theory of demographic dividend and confirmed empirically.
The panel data from World Development Indicators enable us to picture a descriptive
1 In an econometric study, Du (2004) found that population policy, per capita GDP and level of human
capital are decisive factors driving down China’s fertility and empirically identified the different effects
of the three factors.
9
relationship between annual GDP growth rates and TFR levels among countries in the period
of 1960 to the recent years. For those countries and years, in which data are available, annual
GDP growth rates are ranged from -51 to 106 percent. To avoid the complication of
explaining the outliers, we ignore those extreme numbers and only investigate those between
0 and 10 percent, which are assumed to be normal span of annual GDP growth rate.
According to foregoing discussion, the relationship between economic growth rate and
fertility is not a simple linear one but follows a algebraic relationship of a quadratic function.
That is, as TFR declines, the economic growth rate increases first and then declines. In Figure
2, according to the function relations between GDP growth rate and TFR and square term of
TFR, we present the fitted value of annual growth rate of GDP with 95 percent confidence
interval.
3.5
44.5
5
Annual G
row
th R
ate
of G
DP (
%)
0 2 4 6 8Total Fertility Rate
95% CI Fitted values
Figure 2 Empirical Relationship between TFR and GDP Growth
Source: Calculation based on dataset of World Development Indicators
Figure 2 intuitionally pictures reversed U shape pattern of GDP growth rate against
decline in TFR – countries at the lower stage of demographic transition characterized by high
TFR usually suffer poor economic performance; as their TFR levels fall, economic growth
10
speeds up; after a certain point, as TFR further declines and demographic transition enters
later stage characterized by very low TFR, economic growth tends to slow down. Such a
simplified empirical curve is perfectly consistent with the theoretical prediction described
previously. To further examine the statistical significance of the relationship between TFR and
economic growth, by assuming the nonlinear correlation and using foregoing data, we regress
GDP growth rate on TFR and squared term of TFR (Table 1). The regression results show the
reverse U shape relation between GDP growth and TFR by revealing the significantly positive
sign of TFR coefficient and negative sign of squared term of TFR.
Figure 1 Regression Results: the Relation between TFR and Growth
Coefficient Standard error t value P> | t |
TFR 0.6852 0.1133 6.05 0.000
TFR square -0.0736 0.0137 -5.38 0.000
Constant term 3.2359 0.1909 16.95 0.000
Observations 3380
While the more precise explanation based on both economics theory and empirical
evidence requires much more works, the plain fertility-growth nexus here is a sufficient
framework, in which we are now in the position of investigating actual relations between
demographic transition perceived in demography, demographic dividend coined by
demographic economists, and Lewis turning point deduced from development economics,
based on China experiences. In the next section, we will analyze the formation and
anticipated disappearance of demographic dividend, and verdict the advent of Lewis turning
point in the process of China’s economic development.
3. The Economic Impacts of Chinese Demographic Transition
In the first two decades after the establishment of People’s Republic of China in 1949,
the economy rapidly recovered and people’s living standard well improved, which pushed
11
China’s demographic transition into second stage. That is, eliminating the abnormal years
between late 1950s and early 1960, mortality substantially fell, birth rate kept at a chronic
height, and as a result natural growth rate was constantly high. TFR had kept as high as 6 until
1970s and afterwards it declined dramatically. However, the fastest decline of TFR had
happened before the one child policy formally implemented. TFR dropped by 3.5 percentage
points, from 5.8 to 2.3 in the decade of 1970 to 1980, while it dropped by 0.5 to 0.7
percentage point in the entire period between 1980, in which the central government
announced a compulsory population control policy, and the present day, when TFR is agreed
to be 1.6 to 1.8, which are both well below replacement level. Such a fact shows that the
orderly switchover of major stages of demographic transition is primarily the result of
economic growth and social development. In the period of demographic transition from
second to third stages, population at working ages grows faster than dependent population and
therefore proportion of working age population becomes larger and larger, which has released
demographic dividend and upgraded economic growth rate to a level above steady state.
Although the population dependence ratio – namely, the ratio of dependent population
aged 14 and younger and 65 and older to working age population aged 15 to 64, declined as
early as in the mid of 1960s, the substantial increase of working age population and its share
in total population, associated by dramatic fall of population dependence ratio, started in the
mid of 1970s (Figure 3). Such a favorable age structure of population has been translated into
demographic dividend that spurred unprecedented performance of economic growth. A series
of publications (for example, Cai, 2008; Cai and Wang, 2005) explain the rationale, process
and empirics of the demographic dividend in the development of dual economy of China.
Those studies also argue that as such a dividend begins diminishing at the stage of
demographic transition characterized by low birth rate, low mortality and low growth rate of
population, the development of the Chinese dual economy has reached its critical period of
time, the Lewis turning point. This paper is an endeavor synthesizing demographic transition,
demographic dividend and Lewis turning point, accounting for the logical and historical
relations between the three concepts and revealing the challenges facing the Chinese economy
in a changing era.
12
0
200
400
600
800
1000
1200
1400
1600
1950
1955
1960
1965
1970
1975
1980
1985
1990
1995
2000
2005
2010
2015
2020
2025
2030
2035
2040
2045
2050
Po
pu
latio
n (
Millio
n)
0-14 15-64 65+
Figure 3 Changing Trends of Age Structure of Population
Source: United Nations, 2009
The prediction of China’s population and its age composition by United Nations (2009),
shown in Figure 3, is a scenario of medium variant and consistent with the 2000 census, the
2005 sample census, and updated (to 2008) estimates of the subsequent trends in fertility,
mortality and international migration. This prospect is by and large identical to those made by
various Chinese units. According to this prediction, total population of China is going to reach
its peak in 2030 and amount to 1.46 billion. Much sooner than that, the working age
population aged 15 to 64 will reach its peak, 998 million, in 2015. Although such prospects
can be in time obtained from public source, it is obvious that they are unknown information
not only for the public but also for many of economists. To acquaint oneself of the changing
trends of Chinese population is definitely required for the scholars who are studying and
speaking of the future of the Chinese economic development, and influencing the public.
Further examining the predicted results shown in Figure 3, one can find that the growth
speed of working age population has been faster than that of total population in the period of
1970 to 2010, and the trend will turn to the other way round – namely, the age structure of
Chinese population will be no longer evolved to be productive. As agricultural share of labor
force declines over time, the current demand of urban sectors for labor has been sufficiently
met by rural-to-urban migration. As an important part of the reform, opening-up and growth
in the past 30 years, the massive migration has drawn worldwide attention and been
13
recognized as the largest movement of population in peacetime of human history (Roberts et
al., 2004). In the recent years, as a result of long-playing adjustment corresponding to
migration, there have occurred two new phenomena to the labor relocation. On the one hand,
the migrant workers become further unwanted as mechanization of agricultural production
has increasingly accelerated. On the other hand, the demand of urban sectors for migrant
workers becomes more and more rigid and indispensable, as a result of shrinkage of urban
local labor force (Cai, 2010).
According to another prediction, which takes into account the impacts of rural-to-urban
migration (Hu, 2009), by 2015, the amount of incremental working age population in urban
area will be less than that of reduced working age population in rural area (Figure 4), the
same conclusion drawn by the above cited prediction. That implies that without substantial
enhancement of wages and other incentives the migrant workers will not fill up the gap
vacated by rapid reduction of urban labor force. In the course of urbanization, those migrant
workers and their accompanying family members, who live in urban area for more than 6
months, are already counted as urban residents, the number of statistically recognized rural
residents have been rapidly reduced. The labor market has gradually responded to this
situation, which is embodied in tremendous enhancement of wages of ordinary workers, on
the one hand, and nationwide labor shortage, on the other. According to the definition of
development economics, those phenomena are signals of Lewis turning point.
-12000
-8000
-4000
0
4000
8000
12000
16000
20000
2002
2005
2008
2011
2014
2017
2020
2023
2026
2029
2032
2035
2038
2041
2044
2047
2050
Working population (thousand)
Rural Urban
Figure 4 Changes of Working Age Population of Rural and Urban China
14
Source: Hu (2009)
4. How to Close the “Aging before Affluence” Gap?
The worldwide experiences show that economic growth and social development are
major driving forces of demographic transition, while the implementation of population
policy plays relatively minor role additional to the former. All like China, the Asian
economies such as Korea, Singapore, Thailand and Taiwan, where no compulsory policy has
been enforced, have experienced dramatic decline in fertility, from high TFR similar to China
in 1950s to as low as below replacement level in 1990s. Even in India, where economic
growth and social development have not performed as good as that in China and those
economies, therefore demographic transition has been relatively lagging behind, fertility has
also been declining by following the similar path (Lin, 2006).
In spite of its unprecedented economic growth in the past 30 years, due to its late outset
comparing to the Asian economies above mentioned, China entered into the new stage of
demographic transition at relatively low income per capita, which is characterized by “aging
before affluence”. In 2000, the proportion of ages 65 and over in total China’s population was
6.8 percent, identical to the world average, whereas China’s per capita GNI was only 17.3
percent of the world average based on official exchange rate and 56.3 percent of the world
average based on purchasing power parity. While one must admit that the strictly
implementation of one-child policy is an accelerator pushing down the fertility, the
demographic transition, in the final analysis, is the result of outstanding economic growth and
social development (Du, 2004). The difference of income level between China and developed
countries is therefore the root cause of the existing “aging before affluence” gap.
Most developed countries are facing the challenges of population aging to sustainable
economic growth and pension insurance scheme, and the efforts and effects vary from country
to country. However, the developed countries, being at high-income level, technological
innovation frontier, and therefore with increasing productivity, have well dealt with the
challenges and averted the old age crisis so far. Based on the experiences, one can be
15
confident that the key for China to tackle the challenges of shrinking working age population
and enhancing older age population is to sustain its fast economic growth and quick increase
in per capita income. In other words, demographic transition, thus population aging, is an
irreversible process, which cannot be stopped even if one-child policy is abolished. The
already formed “aging before affluence” gap can only be narrowed and eventually closed
through catching up with the developed countries, which put China in the ranks of high
income.
As the Chinese economy ascends itself in the world’s economic pecking order and is
expected to become second largest economy after the United States in 2010, given its slowed
down growth rate of population, per capita GDP level of China will rapidly step up. One
long-term prediction on China’s economic size and per capita GDP conducted by Japanese
Center for Economic Research (JCER, 2007) shows that based on PPP and the constant US
dollar of 2000, China’s GDP volume will reach $17.3 trillion in 2020, $25.2 trillion in 2030,
and $30.4 trillion in 2040. The predicted per capita GDP in the three reference years are $12
thousand, $18 thousand and $22 thousand, respectively. Even more optimistic prediction by
Fogel (2007) expects that China’s total GDP will reach $123.7 trillion in 2040, and based on
the predicted population of 1.46 billion, per capita GDP will be as high as $85 thousand then.
Those two prospects are widely divergent in terms of methodology, data used, assumptions,
and therefore the predicted results. Given the debatable usage of purchasing power parity
GDP term by both researches, their prospects are unlikely to be accepted by Chinese scholars
and officials.
The above-mentioned controversial prospects on China economic volume and per capita
income, however, reveal the same facts that beginning from the second decade of 21st century,
in the stance of second largest economy of the world, China will speed its transformation
from middle-income country to high-income country. Suppose China can maintain the same
pace of or even appreciably lower than the growth rates of both total and per capita GDP
realized in the past 30 years, a significant convergence of wealth between China and
developed countries will be realized. In this regard, the predictions made by foreign
economists reflect the keerect direction and vision of near future, and as the result of the
trends, under the assumption of unchanged demographic transition, the gap between
16
economic development level and population aging will be eventually closed.
In Figure 5, we compare China’s age structure of population with less-developed
countries in the years of 2000 and 2010 to show the characteristic of “aging before affluence”,
whereas we compare China’s age structure of population with more-developed countries in
the years of 2020 to 2030 to show how the “aging before affluence” gap can be narrowed
down and finally closed. This shows that the fundamental avenues to dealing with population
aging in the post Lewis turning point era are threefold: (1) tap the potential of remaining
demographic dividend, (2) create second demographic dividend, and (3) find new sources
sustaining long-run economic growth.
2000
12 10 8 6 4 2 0 2 4 6 8 10 12
0-4
15-19
30-34
45-49
60-64
75-79
90-94
China Less developed regions
2010
12 10 8 6 4 2 0 2 4 6 8 10 12
0-4
15-19
30-34
45-49
60-64
75-79
90-94
China Less developed regions
2020
12 10 8 6 4 2 0 2 4 6 8 10 12
0-4
15-19
30-34
45-49
60-64
75-79
90-94
China More developed regions
2030
12 10 8 6 4 2 0 2 4 6 8 10 12
0-4
15-19
30-34
45-49
60-64
75-79
90-94
China More developed regions
Figure 4 Narrowing the “Aging before Affluence” Gap by Economic Catching up
Source: United Nations, 2009
5. Conclusion and Policy Implications
The analysis on China’s demographic transition and its economic impacts convince us of
the close relation between demographic transition and economic development, and thus the
conclusion of diminishing demographic dividend and ongoing Lewis turning point is pursuant
17
with both economic theory and empirical evidence. Even the scorching debates around those
judgments, to some extent, clue to that they are not solely academic discussions, that they also
have important implications to policy decision.
The arrival of Lewis turning point is vital milestone for a developing economy, because
only does it pass through this point that the marginal productivity of labor in traditional
(agricultural) sector begins to converge that in modern (non-agricultural) sectors. When
eventually the productivity gap among sectors disappears, the economy as a whole receives
its commercial point and the long-standing dual economy terminates. Therefore, the advent of
Lewis turning point is not dismal news at all. In contrast, being cognizant of this turning point
has not only theoretical meaning but also practical significance. That is, it has important
implications for governments in development policy formation, for enterprises in
decision-making, and for individuals in adjustment in the face of labor market changes. In
short, the sustainability of economic growth relies heavily on a sound response to the
challenges brought forward by the new stage of development. What follows we examine the
potential sources of China’s sustained economic growth in post-Lewis turning point period,
based on experiences of foregoer economies.
First, there still is potential of the existing demographic dividend to exploit in the short
run. Let us first divide demographic dividend into two types: first demographic dividend and
second demographic dividend. While the first demographic dividend can be defined as the
scenario where labor supply is adequate and savings rate is high, thanks to the increasing
magnitude and proportion of working age population, the second demographic dividend can
be seen as the new motivation of savings resulted from precautions for the rainy day of aging
and new supply of human capital supposedly brought about by postponing retirement age and
extending education and training (Cai, 2009). The exploitation of first demographic dividend
has been manifested in labor transformation from agricultural to secondary and tertiary
sectors, accompanied by speedy urbanization. By 2009, urban residents who lived in cities for
6 months or longer reached more than 600 million, accounting for 46 percent of Chinese
population. While according to the definition, part of longer stayed rural-to-urban migrants
were counted as urbanites, because they do not legitimately have urban hukou and lack access
to public services, of which urban residents with local hukou are inherently privileged, they
18
are not fully urbanized.
More specifically, because migrant workers and their accompanied families still expect
to return home in rural area, periodically during the Chinese New Year period, and
permanently after they are not needed by the urban labor market, (1) their supply of labor is
not durative, (2) their consumption behavior is still rural, (3) they are not planned for the
utilization of urban infrastructure, and (4) they have no incentives to contribute to the social
security scheme, particularly to the fully-funded pension system. In 2007, the proportion of
urban residents of total Chinese population, which includes those migrant workers who live in
cities for more than 6 months, was 45 percent, whereas the proportion of population who have
formal urban hukou was only 33 percent, which leaves an ullage of 12 percentage points gap
between nominal and actual urbanization rates. By transforming farmers-turned workers to
migrants-turned citizens through deepening the hukou reform and equalizing public services
to all citizens, a more complete urbanization can make full use of the first demographic
dividend (Cai, 2010).
Second, there is potential of second demographic dividend to tap in the medium run. An
aging society can also possess advantage of population – namely second demographic
dividend, as long as necessary institutional conditions are provided with (Lee and Mason,
2006; Cai, 2009). As one of the important causes for population aging, the increase of life
expectancy – people live longer and healthier, is the foundation to generate the second
demographic dividend. This kind of demographic dividend includes three major sources. A
first source comes from the need of old-age supports and supply of pension institutions. As
long as there is fully funded pension scheme instead of pay-as-you-go or family support
system, the older and working longer workforce will have larger incentive to accumulate
assets, and then high savings rate can be maintained by investing in capital market (Lee and
Mason, 2006). A second source comes from the expansion of education resources. As the
quantity of youth population reduces and its proportion in total population declines, the
capacity of working age population maintaining population at school relatively enhances,
which is a window of opportunity for extending education and training and hence accumulate
human capital. A third source comes from expansion of labor force participation. Extension of
retirement age is a major measure to enlarge workforce and alleviate the burden of old-age
19
supports, which is widely employed in developed countries. For China, the obstacle of
extending retirement age is that the working elderly are usually less educated and thus
unwanted by employers. Before this situation is changed, extension of retirement age may put
them in a vulnerable circumstance. Therefore, the exploitation of such a dividend should take
a gradual approach through expanding education, especially that at senior high school level
liberal education and occupational education, and on-job-training, particularly for migrant
workers.
Third, there is a great need to obtain a brand new driving force for the sustainable
economic growth through growth pattern transformation in the long run. Assuming scarcity of
labor and thus diminishing return to capital, the neoclassical theory of growth developed
based on western experiences argues that the only way to maintain the sustained economic
growth is to enhance the contributive share of total factor productivity (TFP) to it (for
example, Solow, 1956). Based on this basic assumption, a handful of studies doubted the
existence of East Asian miracle and, if there is, its sustainability (for example Young,1992;
Krugman, 1994). In reality, thanks partially to the characteristics of unlimited supply of labor
and partially to felicitous economic policies favorable for exploiting demographic dividend,
the East Asian economies had long averted the crisis of diminishing return of capital. As some
major economies passed through their Lewis turning points1, at the same time, the newly
entered stage of demographic transition no longer provided first demographic dividend, those
economies represented by Japan and the Four Tigers transformed their economic growth
pattern from capital and labor driven to more TFP driven. Those experiences indicate that as
first demographic dividend diminishes and Lewis turning point passes through, driving forces
of the Chinese economic growth will be eventually transformed to a reliance on technological
advancement and productivity enhancement. The assurance of changed stage of development
requires to speed up the pace of growth pattern transformation.
1 It is commonly believed that the Japanese economy in 1960 and Korean and Taiwan economies
reached their Lewis turning point, respectively (for example Minami, 1968; Bai, 1982).
20
Reference
Bai, Moo-ki (1982) The Turning Point in the Korean Economy, Developing Economies, No.2,
pp. 117-140.
Cai, Fang (2004) The Consistency of China's Statistics on Employment: Stylized Facts and
Implications for Public Policies, The Chinese Economy, Vol. 37, No. 5 (September-October),
pp. 74 – 89.
Cai, Fang (2008a) Lewis Turning Point: A Coming New Stage of China’s Economic
Development, Social Sciences Academic Press (China).
Cai, Fang (2008b) Approaching a Triumphal Span: How Far Is China Towards its Lewisian
Turning Point? UNU-WIDER Research Paper No. 2008/09.
Cai, Fang (2009) Future Demographic Dividend – Tapping the Source of China’s Economic
Growth, China Economist, No. 21 (July-August).
Cai, Fang (2010) From Farmers-turned Workers To Migrants-turned Residents: Urbanization
In Next Decades, Chinese Style, International Economic Review, No. 2.
Cai, Fang and Dewen Wang (2005) China’s Demographic Transition: Implications for Growth,
in Garnaut and Song (eds) The China Boom and Its Discontents, Canberra: Asia Pacific Press.
Cai, Fang and Meiyan Wang (2008) A Counterfactual Analysis on Unlimited Surplus Labor in
Rural China, China & World Economy, Vol.16, No.1, pp.51-65.
Caldwell, John C. (1976), “Toward a Restatement of Demographic Transition Theory”,
Population and Development Review, 2: 321-366.
Du, Yang (2004) The Formation of Low Fertility and Its Impacts on Long term Economic
Growth in China, The World Economy, No. 12.
Fogel, Robert W. (2007) Capitalism and Democracy in 2040: Forecasts and Speculations,
NBER Working Paper, No. 13184.
Hu, Ying (2009) Predictions on Working Age Population of Rural and Urban China,
unpublished memo.
Japan Center for Economic Research (JCER) (2007) Demographic Change and the Asian
Economy, Long-term Forecast Team of Economic Research Department, Japan Center for
Economic Research, Tokyo.
21
Krugman, Paul (1994), “The Myth of Asia's Miracle”, Foreign Affairs
(November/December).
Lau, Laurence (2010a) Expansion of Domestic Demand is Fourfold, China News Website, 21